1. Field of the Invention
The present invention relates to an anti-FGF23 antibody which specifically binds to an FGF23 antigen. Furthermore, the present invention relates to an agent for prevention or treatment of mineral metabolic disorders due to excessive production of FGF23 or other causes comprising as an active ingredient the anti-FGF23 antibody. In particular, the present invention relates to an agent for treatment of hypophosphatemic rickets and osteomalachia treatment agent.
2. Background Art
Fibroblast growth factor was first purified from a bovine pituitary gland as a substance that stimulates an increase in fibroblast cell line NIH3T3. Since then, similar proteins have been identified various tissues, and a group of the substances compose a polypeptide family (FGF family). Up until now, 22 proteins have been identified in vertebrates as belonging to the FGF family. With regard to the biological activity of these proteins, not only do they have fibroblast growth activity, but these proteins are also known to have divergent actions such as growth of the mesoblast and the neuroectoderm, and angiogenesis action, and limb bud formation in the developmental stage. FGF is also varied in the gene expression site and expression time. They are often expressed only at certain sites only in the developmental stage or in adults. At least 4 genes encoding the FGF receptor are known, FGFR1, FGFR2, FGFR3, and FGFR4. In addition, with regards to FGFR1, FGFR2, and FGFR3, it is known that there are receptor proteins for each with differing extracellular domains due to differences in splicing. In addition, heparin and heparan sulfate proteoglycan are known to control the action by interaction with FGF and FGF receptors. In addition, there are many, which, due to structural similarities, belong to the FGF family, but whose biological activities and receptor binding properties and the like have not been known. The characteristics of this FGF family have been summarized in a review (see Ornitz, D. et al., Genome biology, 2: 3005.1-3005.12, 2001).
FGF23 (in general, may also be represented as FGF-23) was cloned initially from a mouse by a database search using homology with FGF15 and the PCR method. Further, human FGF23 was cloned by using the sequence homology with mouse FGF23. Human FGF23 is a polypeptide with 251 amino acid residues. In addition, as the secretory signal sequence, an amino acid sequence at amino terminal side up to 24 amino acids is predicted to be cleaved at the time of secretion (see Yamashita, T. et al., Biochem. Biophy. Res. Commun., 277: 494-498, 2000). Next, in research on autosomal dominant hypophosphatemic rickets/osteomalachia (henceforth referred to as ADHR), the mutated gene region in ADHR patients was narrowed down and with advancement in the identification of the responsible gene, a mis-sense mutation in the FGF23 gene was discovered characteristically in ADHR patients (see White, K. E. et al., Nature Genet., 26: 345-348, 2000). With this discovery, there was a strong suggestion that FGF23 was physiologically important in the body. On the other hand, what determined the biological activity of FGF23 was research into neoplastic osteomalachia which is one of the hypophosphatemia rickets and osteomalachia diseases. In this disease, the culprit neoplasm of the disease produces and secretes a liquid disease initiating factor, and it is thought that pathologies such as hypophosphatemia, osteomalachia and the like are caused by the disease initiating factor.
In the search for the disease initiating factor produced by this culprit neoplasm, FGF23 was cloned as a gene which is overexpressed in the tumor. Furthermore, by administering this factor, it was shown that hypophosphatemia and osteomalachia were reproduced (see Shimada, T. et al., Proc. Natl. Acad. Sci., 98: 6500-6505, 2001 and International Publication Number WO02/14504 pamphlet). Based on this research, FGF23 has been shown to be related in the metabolic control related to phosphorus and calcium in the body. In addition, it was suggested that this acts as a systemic factor which expresses its action by circulating in the body. Furthermore, later research also showed that the blood of actual neoplastic osteomalachia patients had a higher value of FGF23 concentration as compared to healthy subjects (see Yamazaki, Y. et al., J. Clin. Endocrinol. Metab., 87: 4957-4960, 2002 and Jonsson, K. B., et al., N. Engl. J. Med., 348: 1656-1663, 2003).
In addition, X-linked hypophosphatemic rickets (henceforth referred to as XLH) is a disease which is known to have a similar presentation as ADHR and neoplastic osteomalachia in terms of clinical findings. In this disease as well, the FGF23 concentration in the blood was shown to be at a high value (see Yamazaki, Y. et al., J. Clin. Endocrinol. Metab., 87: 4957-4960, 2002 and Jonsson, K. B., et al., N. Engl. J. Med., 348: 1656-1663, 2003).
In other words, the cause for vitamin D resistant rickets and osteomalachia which were observed in neoplastic osteomalachia, XLH, and the like had been previously unknown, but the secreted disease causing factor was shown to be FGF23. Furthermore, with regard to other mineral metabolic diseases such as fibrous dysplasia, McCune-Albright syndrome, autosomal recessive hypophosphatemia rickets, and the like, high concentrations of FGF23 in the blood have been reported to be associated with hypophosphatemia and rickets and osteomalachia (See Riminucci, M. et al., J. Clin. Invest., 112: 683-692, 2003; Yamamoto, T. et al., J. Bone Miner. Metab., 23: 231-237, 2005; Lorenz-Depiereux, B. et al., Nat. Genet., 38: 1248-1250, 2006).
From the above report, the condition of having excessive FGF23 in the body has been shown to induce hypophosphatemia and the accompanying rickets and osteomalachia and the like. Furthermore, for chronic renal insufficiency hyperphosphatemia, abnormally high serum FGF23 values have been reported. Excessive FGF23 has been suggested to be possibly related to a portion of the mineral metabolic diseases during renal insufficiency (see Gupta, A. et al., J. Clin. Endocrinol. Metab., 89: 4489-4492, 2004 and Larsson, T. et al., Kidney Int., 64: 2272-2279, 2003). With regard to these diseases induced due to excessive FGF23, suppressing the action of FGF23 or removing FGF23 is thought to be a possible way to treat the diseases. Up to now, anti-FGF23 mouse monoclonal antibody has been reported to be a way to suppress the action of FGF23 (see Yamashita, T. et al., Biochem. Biophy. Res. Commun., 277: 494-498, 2000). When the anti-FGF23 mouse monoclonal antibody 2C3B and 3C1E used in this report were administered to normal mice, the function of the endogenous mouse FGF23 was inhibited, and the phosphorus excretion from the kidney was suppressed. By fluctuating the expression of vitamin D-metabolizing enzyme in the kidney, this was shown to result in increased concentrations for phosphorus and 1 α,25 dihydroxy vitamin D (henceforth referred to as 1,25D) in the serum. Furthermore, repeated administration of anti-FGF23 mouse monoclonal antibody was conducted on Hyp mouse which is a model mouse for XLH which has a high serum concentration of FGF23 and has hypophosphatemia and has bone elongation dysfunction and calcification dysfunction. As a result, in the Hyp mice, a rise in the phosphorus concentration in the blood was seen, and in addition, there were improvements in bone elongation dysfunction and calcification dysfunction. From these results, the use of FGF23 action suppressing antibody was thought to be appropriate as a medicine for FGF23 excess diseases. However, the 2C3B and 3C1E antibodies used in this report are mouse-derived antibodies. Mouse antibodies which are recognized as foreign by human host initiates a so-called “human anti-mouse antibody” (in other words HAMA) response, and there may be situations where serious side-effects are seen (see Van Kroonenbergh, M. J. et al., Nucl. Med. Commun. 9: 919-930, 1988).
In order to avoid this type of problem, one approach was to develop a chimera antibody (see European Patent Application Publication Number 120694 Specification and European Patent Application Publication Number 125023 Specification). Chimera antibodies include a portion of antibody derived from 2 or more species (for example, variable region of the mouse antibody and the constant region of the human antibody and the like). The advantage of this type of chimera antibody is that the binding to the antibody which was the characteristic of the original mouse antibody is maintained, but on the other hand, “a human-anti chimera antibody” (in other words “HACA”) response is still induced (see Bruggemann, M. et al., J. Exp. Med., 170: 2153-2157, 1989).
Furthermore, a recombinant antibody has been developed where only a portion of the substituted antibody is a complementarity determining region (CDR) (see British Patent Number GB2188638A specification and U.S. Pat. No. 5,585,089 specification). Using CDR transplant technology, an antibody consisting of mouse CDR, the framework of the human variable region and constant region (in other words “humanized antibody”) was produced (see Riechmann, L. et al., Nature, 332: 323-327, 1988). It has been known that using this method, anti-FGF23 mouse antibody such as 2C3B antibody can be humanized by substituting mouse antibody with a human antibody sequence. However, when humanized, there is the possibility that the affinity to the antigen may be reduced. In addition, for the current treatment of hypophosphatemia rickets in XLH and the like, the main method is periodic oral administration of Vitamin D formulation and phosphoric acid. However, there is the problem that the patients are forced to have a substantial burden due to the size of each dose and the dosage frequency per day. Therefore, in order to lessen the burden on the patients and their families, a hypophosphatemia treatment drug which shows a sustained raising action for serum phosphate concentration and serum 1,25D concentration is desired in order to extend the time between doses.